Eleanor Drage's new book, What If We Got AI Right, presents a refreshingly optimistic yet pragmatic vision for the future of artificial intelligence, arguing that the much-feared AI apocalypse can be avoided through deliberate, human-centered design and policy.
A Practical Alternative to Doom
Drage, a senior research associate at the University of Cambridge's Leverhulme Centre for the Future of Intelligence, pushes back against the dominant narrative that AI will inevitably lead to widespread job loss, bias, or existential risk. Instead, she offers concrete examples of how AI is already being used for good, from healthcare diagnostics to climate change modeling. According to Drage, the key is not to slow down AI development but to steer it toward socially beneficial outcomes.
The book is structured around case studies of successful AI implementations, such as an AI system that helps radiologists detect breast cancer earlier and a chatbot that provides mental health support in underserved communities. Drage emphasizes that these successes are not accidental; they result from inclusive design processes and robust regulatory frameworks.
The Role of Regulation and Ethics
A central theme is that regulation is not the enemy of innovation. Drage argues that clear ethical guidelines and government oversight can actually accelerate responsible AI adoption. She cites the European Union's AI Act as a promising model, though she notes it must be paired with enforcement and adaptability. The book also explores the importance of diversity in AI development teams, showing that homogeneous groups are more likely to produce biased or harmful systems.
Drage does not shy away from the challenges. She dedicates a chapter to the risks of AI in surveillance and autonomous weapons, but frames these as problems we can solve with political will and public pressure. She quotes a leading AI ethicist: “The future is not written in code; it is written in the choices we make today.”
Impact and Takeaways
The book's most compelling contribution may be its call for a new mindset: instead of asking what AI will do to us, we should ask what we want AI to do for us. Drage provides a toolkit for policymakers, technologists, and citizens to evaluate AI systems and demand accountability. She includes a checklist of questions to ask before deploying any AI, such as: Who benefits? Who might be harmed? How is the data sourced?
Ultimately, What If We Got AI Right is a timely antidote to tech pessimism. It offers a roadmap that is both ambitious and achievable, grounded in real-world examples and rigorous research. For anyone concerned about the direction of AI, Drage's book is a must-read—not because it promises easy answers, but because it shows that the future is still ours to shape.
